1. Field of the Invention
This invention is related to the field of computer systems and, more particularly, to the sharing of non-distributed, random-access storage.
2. Description of the Related Art
Enterprise computing systems are increasingly using configurations such as storage area networks (SANs), network attached storage (NAS), and other centralized storage mechanisms to simplify storage, improve availability, and handle escalating demands for data and applications. The storage area network (SAN) model places storage on its own dedicated network. This dedicated network most commonly uses Fibre Channel technology as a versatile, high-speed transport. The SAN may include one or more storage hosts that provide a point of interface with local area network (LAN) users and may also include one or more fabric switches, SAN hubs, and/or other intermediate entities to accommodate a large number of storage devices. The hardware (e.g., switches, hubs, bridges, routers, cables, etc.) that connects servers to storage devices in a SAN is referred to as a “disk fabric” or “fabric.” The SAN fabric may enable server-to-storage device connectivity through Fibre Channel switching technology to a wide range of servers and storage devices.
As the size and complexity of networked computing environments are increasing, so too are the demands placed on backup solutions. Tape devices have traditionally been used as a high-capacity backup medium. To permit more efficient use of tape resources, backup solutions (e.g., NetBackup™ with Shared Storage Option) from Symantec Corporation have provided for the sharing of tape devices in a networked backup environment. Additionally, it may be desirable to use available disk-based storage (e.g., in a SAN) for backup, either as a final backup destination or as an intermediate location for staging the data to tape.
One prior approach has used a single server and its associated storage to provide backup for other computer systems on a network. By providing a single point of execution, this configuration may be subject to media contention issues and other limitations in system resources (e.g., CPU cycles, available memory, I/O bandwidth, etc.). By providing a single point of failure, this configuration may be vulnerable to errors in both hardware (e.g., CPU, memory, network and storage interfaces, power supply, etc.) and software (e.g., operating system, drivers, backup software, etc.).
Distributed file systems or clustered file systems typically support the sharing of persistent storage resources among multiple hosts over a network. Examples of such file systems include Network File System (NFS) and Common Internet File System (CIFS). However, such file systems may not be usable in all environments due to limited hardware support, limited scalability, and/or undesirable overhead.